Possible release notes content: This feature adds two new election policies. Said policies are: org.jboss.ha.singleton.HASingletonElectionPolicySimpleQuorum org.jboss.ha.singleton.HASingletonElectionPolicyQuorum 'Quorum' implies that election of master node does not happen unless there is minimum number of nodes met ( in cluster ). This applies to both, new nodes joining and nodes falling out of cluster - either due to network split or nodes just being shut down. 'Quorum' is configurable parameter in both policies( setQuorum ). The HASingletonElectionPolicyQuorum has only quorum configurable. The HASingletonElectionPolicySimpleQuorum has an offset, just like org.jboss.ha.singleton.HASingletonElectionPolicySimple . Example setup: <bean class="org.jboss.ha.singleton.HASingletonElectionPolicyQuorum" name="ExampleHASingletonElectionPolicy"> <property name="quorum">2</property> </bean> <bean class="org.jboss.ha.singleton.HASingletonController" name="ExampleHASingletonController"> <annotation> @org.jboss.aop.microcontainer.aspects.jmx.JMX(name="atest:service=ExampleHASingletonController", exposedInterface=org.jboss.ha.singleton.HASingletonControllerMBean.class, registerDirectly=true) </annotation> <property name="clusterPartition"> <inject bean="ExampleHAPartition"/> </property> <property name="electionPolicy"> <inject bean="ExampleHASingletonElectionPolicy"/> </property> <property name="target"> <inject bean="ExampleHASingletonMBean"/> </property> <property name="targetStartMethod">startSingleton</property> <property name="targetStopMethod">stopSingleton</property> <property name="targetStopMethodArgument">true</property> </bean>